IAEA-TECDOC-1499
Intercomparison of techniques for inspection and diagnostics of heavy water reactor pressure tubes
Flaw detection and characterization
IAEA-TECDOC-1499
Intercomparison of techniques for inspection and diagnostics of heavy water reactor pressure tubes
Flaw detection and characterization
The originating Section of this publication in the IAEA was:
Nuclear Power Technology DevelopmentSection International Atomic Energy Agency
Wagramer Strasse 5 P.O. Box 100 A-1400 Vienna, Austria
INTERCOMPARISON OF TECHNIQUES FOR INSPECTION AND DIAGNOSTICS OF HEAVY WATER REACTOR PRESSURE TUBES:
FLAW DETECTION AND CHARACTERIZATION IAEA, VIENNA, 2006
IAEA-TECDOC-1499 ISBN 92–0–105606–0
ISSN 1011–4289
© IAEA, 2006
Printed by the IAEA in Austria
FOREWORD
Nuclear power plants with heavy water reactors (HWRs) comprise nine percent of today’s operating nuclear units, and more are under construction. Efficient and accurate inspection and diagnostic techniques for various reactor components and systems are an important factor in assuring reliable and safe plant operation.
To foster international collaboration in the efficient and safe use of nuclear power, the IAEA conducted a Coordinated Research Programme (CRP) on Inter-comparison of Techniques for HWR Pressure Tube Inspection and Diagnostics. This CRP was carried out within the frame of the IAEA Department of Nuclear Energy’s Technical Working Group on Advanced Technologies for HWRs (the TWG-HWR). The TWG-HWR is a group of experts nominated by their governments and designated by the IAEA to provide advice and to support implementation of the IAEA’s project on advanced technologies for HWRs.
The objective of the CRP was to inter-compare non-destructive inspection and diagnostic techniques, in use and being developed, for structural integrity assessment of HWR pressure tubes. During the first phase of this CRP, participants have investigated the capability of different techniques to detect and characterize flaws. During the second phase of this CRP, participants collaborated to detect and characterize hydride blisters and to determine the hydrogen concentration in Zirconium alloys. The intent was to identify the most effective pressure tube inspection and diagnostic methods, and to identify further development needs.
The organizations that have participated in this CRP are:
• The Comisión Nacional de Energía Atómica (CNEA), Argentina;
• Atomic Energy of Canada Ltd. (AECL); Chalk River Laboratories (CRL), Canada;
• The Research Institute of Nuclear Power Operations (RINPO), China National Nuclear Corporation (CNNC), China;
• Bhabha Atomic Research Centre (BARC), India;
• The Korea Electric Power Research Institute (KEPRI), Republic of Korea;
• The Korea Atomic Energy Research Institute (KAERI), Republic of Korea;
• The National Institute for Research and Development for Technical Physics (NIRDTP), Romania; and
• Nuclear Non-Destructive Testing Research and Services (NNDT), Romania.
This TECDOC reports the results of the collaboration on flaw detection and characterization (Phase 1). A companion publication reports results of the collaboration on characterizing hydride blisters and determining the hydrogen concentration in Zirconium alloys (Phase 2).
The IAEA officer responsible for this publication was J. Cleveland of the Division of Nuclear Power.
EDITORIAL NOTE
The use of particular designations of countries or territories does not imply any judgement by the publisher, the IAEA, as to the legal status of such countries or territories, of their authorities and institutions or of the delimitation of their boundaries.
The mention of names of specific companies or products (whether or not indicated as registered) does not imply any intention to infringe proprietary rights, nor should it be construed as an endorsement
CONTENTS
1. INTRODUCTION...1
1.1. Background, and conduct of this CRP ...1
1.2. Importance of inter-comparison of PT inspection and diagnostic techniques...2
1.3. Participations from the various laboratories...3
1.3.1. Argentinean participation ...3
1.3.2. Indian participation...3
1.3.3. Korean participation ...3
1.3.4. Romania-NIRDTP participation...4
1.3.5. Romania-NNDT participation...4
1.3.6. Chinese participation ...5
1.3.7. Canadian participation...5
1.4. Summary of sample preparation and description...5
1.4.1. Argentinean sample (ARG1)...5
1.4.2. Indian sample (IND1)...6
1.4.3. Korean sample (KOR1)...6
1.4.4. Romanian sample (ROM2)...6
1.4.5. Chinese sample (CHI1) ...6
1.4.6. Canadian samples (CAN1 and CAN2)...7
1.5. Inspection and diagnostic techniques...7
1.5.1. Argentina ...7
1.5.2. India...7
1.5.3. Republic of Korea...7
1.5.4. Romania-NIRDTP...8
1.5.5. Romania-NNDT ...8
1.5.6. Canada ...8
2. FLAW DETECTION AND CHARACTERIZATION ...10
2.1. Flaw description in PT samples ...10
2.1.1. Argentinean sample (ARG1)...10
2.1.2. Indian sample (IND1)...13
2.1.3. Korean sample (KOR1)...15
2.1.4. Romanian sample (ROM2)...17
2.1.5. Chinese sample (CHI1) ...20
2.1.6. Canadian samples (CAN1 and CAN2)...23
2.2. Inspection and diagnostics techniques ...26
2.2.1. Argentina ...26
2.2.2. India...26
2.2.3. Republic of Korea...28
2.2.4. Romania-NIRDTP...28
2.2.5. Romania-NNDT ...30
2.2.6. Canada ...33
3. DISCUSSION OF RESULTS ...42
4. CONCLUSIONS AND RECOMMENDATIONS...44
4.1. Conclusions ...44
4.2. Recommendations ...44
APPENDIX I. GLOSSARY ...47 REFERENCES ...55
CONTRIBUTORS TO DRAFTING AND REVIEW ...57
ATTACHED CD-ROM:SAMPLE SUMMARY REPORTS OF PRESSURE TUBE SAMPLES
1. INTRODUCTION 1.1. Background, and conduct of this CRP
The pressure tubes (PTs) of heavy water reactors (HWRs) operate in a high-temperature high- pressure aqueous environment and are subjected to fast neutron irradiation. These conditions lead to degradations in the PTs with respect to i) dimensional changes (creep and growth), ii) deterioration of mechanical properties (hardening and embrittlement) thereby reducing their flaw tolerance, iii) the growth of existing flaws, which were too small or ‘insignificant’ at the time of installation, and iv) initiation and growth of new flaws. The PTs, which are made of Zirconium alloy (Zircaloy-2 or Zr-2.5% Nb), also undergo corrosion in aqueous environment during service. This releases hydrogen, a part of which gets absorbed in the PT material. The absorbed hydrogen is responsible for the two most commonly observed degradation mechanisms that can limit the life of a PT: delayed hydride cracking (DHC) and hydride blister formation and cracking. In HWRs with Zircaloy-2 PTs, an additional mechanism in the form of accelerated oxidation and hydriding can lead to its premature retirement from service, when the oxide layer thickness on its inner surface exceeds the threshold limit.
The PT integrity is central to the safety of HWRs. In order to ensure the PT integrity at all times during their service, they are periodically examined by non-destructive examination (NDE) techniques. These inspections provide valuable information on the presence or absence of flaws in the PTs and their characteristics to designers, plant operators and regulatory authorities for fitness-for-service assessment. NDE methodologies for in-service inspection of PTs have been developed and implemented on in-service inspection tools in several countries operating HWRs.
The International Atomic Energy Agency (IAEA) convened two meetings to provide a forum for technical discussion and exchange of information and experience on HWR PT maintenance and replacement, and to provide recommendations to the IAEA for future cooperative activities in this area. These recommendations were considered at the first meeting of the International Working Group on Advanced Technologies for Heavy Water Reactors (IWG-HWR — now called the Technical Working Group on Advanced Technologies for HWRs), held in Vienna, in June 1997. At this meeting, the IWG-HWR advised the IAEA to convene a consultancy to formulate a coordinated research programme (CRP) to address the application of advanced technologies to the safety and reliability of HWRs. The consultancy held at the IAEA in July 1999 formulated the CRP entitled Inter- comparison of Techniques for Pressure Tube Inspection and Diagnostics. The CRP was divided in two Phases: Phase 1 dealing with ‘Flaw characterization by in situ NDE techniques’ and Phase 2 with ‘Blister characterization by in situ NDE techniques’ and
‘Determination of hydrogen in Zirconium alloy components’.
Phase 1 of the CRP was conducted in a round-robin manner. The participating laboratories prepared PT samples containing artificial flaws resembling real defects of concern. The flaws on the outside surface were hidden by a cover to facilitate blind testing. All samples had to be inspected from the inside surface, as in real conditions. The samples, after examination by
effective techniques for blister characterization and determination of hydrogen concentration in PTs.
This TECDOC presents to the HWR community the capability of NDE techniques, existing and under development, for flaw detection and characterization that were examined during Phase 1 of this CRP. A companion TECDOC presents the results for hydride blister characterization and determination of hydrogen concentration in PTs achieved during Phase 2 of this CRP.
Most of the techniques examined in Phase 1 are well established and are regularly used during in-service inspection of PTs. The inter-comparison of these techniques provides a platform for identifying a particular NDE technique (or a set of techniques), which is more accurate and reliable as compared to others for a specified task. The CRP also witnessed some new NDE methodologies, which can be implemented on in-service inspection tools. These new techniques could complement the existing ones to overcome their limitations, thereby improving the reliability and accuracy of in-service inspection. Finally, this TECDOC identifies future areas of research and development and opens up new avenues for technical collaboration within the HWR community to overcome the challenges faced in PT inspection and diagnostics.
1.2. Importance of inter-comparison of PT inspection and diagnostic techniques
Periodic in-service inspection of PTs by NDE techniques is one of the regulatory requirements for HWRs. The objectives of these inspections are to detect, locate and characterize the flaws in PTs and provide information for fitness-for-service assessment. To assure the structural integrity of PTs at all times during service, it is essential to employ NDE techniques that can reliably detect all the ‘significant’ flaws and characterize them accurately.
In order to assess the effectiveness of these techniques for their intended purpose, it is important that they are periodically subjected to ‘blind tests’ on PT samples containing known flaws. This CRP on ‘Inter-comparison of Techniques for Pressure Tube Inspection and Diagnostics’ gave the opportunity for the participating laboratories to prepare their own PT samples containing flaws and carry out blind tests on PT samples prepared by others.
Because these were blind tests, the inspection teams, which examined PT samples, were unaware about the number, location and characteristics of flaws in the samples. This situation is similar to the one faced during in-service inspection of PTs in operating HWRs. The results of examination on PT samples can thus be directly correlated with the effectiveness of NDE techniques for detection and characterization of flaws. A good detectability and accurate characterization would strengthen the confidence in the technique(s) employed, while poor detectability and inaccurate characterization would give a feedback to the laboratory that the existing technique(s) need(s) improvement(s) or new ones should be developed.
The inter-comparison of NDE techniques based on the results of investigation of PT samples highlights the most reliable and accurate NDE method (ultrasonic, eddy-current or a combination of both) and also a specific technique for that NDE method (time-of-flight monitoring, amplitude monitoring, C-scan image, etc.) for detection and characterization of various kinds of flaws encountered in PTs. This information is useful for the HWR community to improve the tools being used for PT inspection and diagnostics, by modifying the existing techniques or adapting new ones, so that the inspection is carried out in the most effective manner. The inter-comparison of NDE techniques also helps in identifying those areas where none of the existing techniques are reliable or effective to the desired level. It
1.3. Participations from the various laboratories
The ultrasonic testing (UT) and eddy-current testing (ET) terms are defined in the glossary in Appendix I.
1.3.1. Argentinean participation
The CNEA produced flaws on a piece of PT (ARG1) provided by India. The inspection team performed UT examinations of the samples from all participating countries.
The CNEA hosted the second Research Coordination meeting (RCM) in Buenos Aires, Argentina, in May 2001. Participants from Argentina, Canada, China, India, Rep. of Korea, Romania, and the IAEA attended this meeting. The purpose of the meeting was to review progress of Phase 1 with respect to sample preparation, transfer and inspection, and prepare an action plan for Phase 2.
During this CRP, laser-optical techniques were also tested, investigating their capability as a NDE technique for PT inspection.
1.3.2. Indian participation
Artificial flaws that duplicate real defects of concern were machined in an approximately 500 mm long PT sample. Blank PT pieces of similar length were provided to Argentina and Peoples Rep. of China, at no cost, so that they could embed flaws in them and then put these samples into circulation.
India examined PT samples from Argentina, Canada, China, Rep. of Korea, Romania and its own sample. The examination was carried out in immersion conditions using a specially- designed scanning system. An inspection head containing UT and ET probes was also designed and fabricated. The flaws in PT samples were fully characterized with respect to their location, orientation, size and type. Conventional and advanced UT techniques were employed for this purpose. A detailed sample inspection report was prepared for each sample and sent to the originating laboratories.
For the Indian PT sample (IND1), the ‘flaw truth’ (true flaw dimensions) was determined for all the machined flaws. The IND1 sample inspection reports from all the participating laboratories were analysed and a detailed sample summary report was prepared. India provided an extensive analysis of the inspection results, which was adopted in this TECDOC, to help identify the most effective and reliable NDE methods for detection, location and sizing of various types of flaws in PTs. India prepared the draft of the sections of this TECDOC on
‘Background and conduct of this CRP’ and ‘Importance of inter-comparison of PT inspection and diagnostics techniques’. India also prepared the sample summary report for the Chinese sample (CHI1).
1.3.3. Korean participation
The Rep. of Korea also examined PT samples from Argentina, Canada, China, Romania, and its own sample. The examinations were carried out using a custom-made scanning system equipped with an ET send-receiver probe. The artificial flaws in PTs were characterized with respect to their locations, size and shape. Detailed sample inspection reports were prepared for each sample and sent to the originating laboratories.
1.3.4. Romania-NIRDTP participation
In 2000, the IAEA approved the participation of the National Institute of Research and Development for Technical Physics (NIRDTP) in this CRP. The NIRDTP conducted their activities based on annual research contracts between IAEA and the NIRDTP. The NIRDTP accomplished all the contracted tasks as described below:
(a) Designed, built and employed an ET send-receive transducer with rotating magnetic field (RMF) for the inspection of the PT samples [1], [2], [3], [4].
(b) Completed model and solved the forward problem [5], [6], [7].
(c) Solved inverse problem using numerical code for evaluation of discontinuities in PT samples [8], [9].
(d) Examined all samples circulated during the CRP, except ROM2, and prepared the corresponding sample inspection reports. The shorter length of this sample as well as the presence of a material transition prevented an inspection with the available ET system.
1.3.5. Romania-NNDT participation
Romania-NNDT designed and constructed specialized equipment, the ISIS-UT&ET Laboratory Multichannel Inspection System for examination of PT samples. It is a complex laboratory system to simulate fuel channel inspections. The functional structure of this system integrates all applicable flaw detection and sizing UT and ET methods addressed by the Periodic Inspection Standard CAN/CSA N285.4 [10], referred in the rest of the document as the Canadian Standard. For this system, Romania-NNDT developed dedicated software for data acquisition and analysis.
The ISIS-UT&ET equipment was completed and improved by implementation of the new inspection methodology and techniques elaborated in the framework of this CRP.
Romania-NNDT examined all samples circulated during the CRP (ARG1, CAN1/2, CHI1, IND1, KOR1 and ROM2). The objective of these inspections was to detect size and characterize the intentional and non-intentional flaws existing in the samples, applying the examination methodology, calibration requirements, and recording criteria specified in the Canadian Standard. The reporting of the detected indications was made according to the requirements of the Canadian Standard, and also respected the “Terms of Reference”
established in the framework of the CRP.
For the PT sample inspections, conventional (routinely used) inspection methods as well as new advanced methods were applied in order to test their capability, accuracy and reliability.
All sample inspection reports completed by Romania-NNDT during this CRP contain very detailed description of the inspection results, and extensive appendices with the collected data for all reportable indications (including the non-intentional indications).
Romania-NNDT designed and constructed its own sample (ROM2) introducing flaws that permit the characterization of the intrinsic performances of the inspection methodology applied by the investigating laboratories.
Romania-NNDT attended and participated actively in all the RCMs.
1.3.6. Chinese participation
China participated in the CRP up to April 2004, at which time China then announced their withdrawal due to a budget cut within their organization. Before that, China had prepared a sample (CHI1) from a piece of PT provided by India. After their withdrawal, China granted continuation of the circulation of the sample as originally planned. Canada provided the ‘flaw truth’ for this sample and India wrote the sample summary report. Before its withdrawal, China had examined samples from two countries (Canada and Rep. of Korea) and was able to provide the corresponding sample inspection reports to the originating laboratories for inclusion in the sample summary reports. It was agreed not to assess China’s NDE capability in this TECDOC due to limited inspection results.
1.3.7. Canadian participation
Canada led Phase 1 of the CRP. Canada prepared jointly with Argentina and presented to the TWG-HWR in Vienna (June 2004) a case for an extension of the CRP (until end of 2005);
this request was approved. Canada assisted in the organization of all meetings and hosted the final RCM.
Canada defined what the reference flaws should be and gave information for representative flaws and how to manufacture them. Canada provided two samples to cover a wide range of flaws, helped coordinate the rotation of all samples and resolve other issues as they occurred all along the CRP.
Canada performed on all samples UT examinations of all flaws, ET and profilometry examinations of the ID flaws.
Beyond conducting profilometry examinations on their own samples (CAN1 and CAN2) to provide full characterization of all flaws, Canada did the same for the KOR1 and CHI1 samples for the respective sample summary reports to be completed on time.
Canada provided a template and scope for the sample inspection reports as well as recommendations for the sample summary reports. Canada produced the draft assessment/analysis part of this TECDOC.
1.4. Summary of sample preparation and description
All PT samples were made as per the requirements decided upon at the opening meeting of the CRP. All samples were made of un-irradiated Zr-2.5% Nb PT material.
1.4.2. Indian sample (IND1)
Artificial notches and grooves were machined in the IND1 sample in such a manner that they closely resemble the real defects. The PT sample also contained reference notches for calibration as required by the Canadian Standard. The IND1 sample has a total of 17 flaws, 12 on the OD surface and five on the ID surface. These flaws can be categorized in seven categories: i) DHC in axial direction on ID and OD surfaces, ii) DHC in circumferential direction on ID and OD surfaces, iii) DHC, oblique to the tube axis, iv) lap-type or laminar defect near ID, OD and mid-wall v) smooth debris fret, vi) sharp debris fret and vii) bearing- pad fret.
1.4.3. Korean sample (KOR1)
Defects, such as fretting damage and crack-like notches representative of those that can arise in CANDU reactors were introduced into the KOR1 sample. It contained a total of 12 intentional flaws, eight on the OD surface and four on the ID surface. All flaws on the inner and external surfaces were electro-discharge machining (EDM) notches with rectangular and semi-elliptical profiles. A range of flaw sizes and orientations was selected. A few unintentional scratches (handling at Wolsong NPP) were present on internal and external surfaces of the sample.
1.4.4. Romanian sample (ROM2)
The main design objective of the ROM2 sample was to provide a sample containing
“families” of flaws of various dimensions for comparison of the intrinsic performances of the examination methods used by the investigating laboratories, in terms of detection sensitivity, resolution and sizing accuracy. The ROM2 sample consisted of three PT rings containing 60 artificial flaws machined through different mechanical processes: drilling, milling, EDM. The families of machined flaws were radial round and flat bottom holes, axial holes, axial and circumferential holes with round, flat or angled (V-form) channels. The ROM2 sample also contained a number of flaws close to the edge of the PT to evaluate the edge effects, simulating flaws located in the rolled joint region. Some flaws were more representative of real flaws, like the flat, and angled (V-form) ID channels, which simulate debris flaws.
Because of the small dimensions and high number of machined flaws, the ROM2 sample was especially intended to test the performance of UT techniques, but its usefulness for characterising some aspects of the ET techniques was also significant.
The ROM2 sample does not contain reference flaws for calibration according to the Canadian Standard. It was assumed that calibration could be achieved using a separate piece of PT.
1.4.5. Chinese sample (CHI1)
The CHI1 sample contained the four calibration slots (axial and circumferential ID and OD) as per the Canadian Standard. A total of 26 flaws were machined in this sample. Out of these, 21 flaws were on the OD surface, and the remaining five on the ID surface. No flaws were added within wall. The flaws were classified by type (axial flaws, circumferential flaws, flaws with tilt, oblique flaws, lap-type or laminar flaws).
1.4.6. Canadian samples (CAN1 and CAN2)
Both samples (CAN1 and CAN2) contained the four calibration notches as well as the three normal beam calibration slots as per the Canadian Standard. Both samples had multiple flaws of various types on ID and OD surfaces, 17 in CAN1 sample and five in CAN2 sample, as representative as possible to flaws encountered in-service: OD axial and circumferential semi- elliptical (simulating DHC), ID axial debris frets of varying complexity with and without DHC, ID axial and circumferential undercut with and without DHC, ID bearing pad frets of various complexity with and without DHC in a corner and an OD axial scratch (simulating an installation scratch). Replicas of each flaw were taken at the end of the round-robin exercise to obtain a precise characterization of each flaw.
1.5. Inspection and diagnostic techniques 1.5.1. Argentina
Argentina used UT examinations for detection, location and characterization of flaws in the various PT samples. The shear wave pulse echo technique was employed using four transducers, two (in opposite directions) in the axial direction more sensitive to circumferential indications and two (in opposite directions) in the circumferential direction more sensitive to axial indications. The four reference notches from the Canadian Standard (ID axial, ID circumferential, OD axial and OD circumferential) were used for calibration and setting of the detection level.
1.5.2. India
UT examination was carried out using four angle beam shear wave transducers (10 MHz) and two normal beam transducers, one focused on the OD surface (10 MHz) and the other on the ID surface (20 MHz). The techniques employed for flaw detection included angle beam pulse- echo, amplitude drop monitoring during angle beam pitch-catch and normal beam backwall drop. The scanning sensitivity for pulse-echo UT examination was set two to four times above the reference sensitivity. The recording criteria was set at 6dB above the noise level, which is very stringent as compared to conventional criteria of 20% or 50% of reference level. For the amplitude drop techniques (normal and angle beam pitch-catch), any indication showing a drop of more than 5 to 10% of the reference signal was evaluated. Once the flaw was detected, length, width and depth were sized. The flaw length and width were determined by the 6dB drop technique using angle beam and normal beam signals. The flaw depths were determined by time-of-flight using angle beam pitch-catch and normal beam examinations.
Records for the flaws included B-scan images and time-of-flight and amplitude plots.
1.5.3. Republic of Korea
Korea used an ET method for flaw detection and characterization. The send-receiver ET transducer was primarily designed to detect ID flaws in the PT samples. This ET transducer consists of two coils. One coil acts as a sender, and the other coil as a receiver. The primary frequency of 60 kHz was used for detection of the ID flaws. A second frequency of 20 kHz
50% above the reference sensitivity. Any signals over reference level was recorded and evaluated.
1.5.4. Romania-NIRDTP
Romania-NIRDTP examined PT samples by ET using a RMF send-receive probe. The experimental and theoretical aspects of this technique need to be further refined.
The emission part of this transducer is made of three rectangular coils, wound with 120º angle between them, star connected and driven by a three-phased alternative current. The magnetic field created by the three coils is a rotating field of a frequency equal to that of the three-phase current. The receiver is made of an array of 24 identical coils equally spaced circumferentially.
Three programmable function generator cards supply the circuit with the three-phase alternative current, as well as the reference for synchronous detection. The signals are amplified and applied to the transmitter creating a RMF. The 24 receiver coils are consecutively interrogated and acquired by a PC.
The RMF transducer and the control equipment were integrally designed and constructed at NIRDTP. The equipment allows to clearly detect a 0.1 mm wide and 0.1 mm deep slot placed on the ID or on the OD of a PT with signal to noise ratio better than 3:1.
1.5.5. Romania-NNDT
The PT samples of this CRP were examined using various UT techniques. The inspection head contains 10 MHz transducers for angle beam (45º shear waves, double skip/W pattern — four angle beam transducers, one pair looking axially and the other looking circumferentially) and normal beam transducers (25 MHz focused on ID, 25 MHz focused on OD, 25 MHz focused on mid-wall, 10 MHz focused on OD, and 10 MHz focused on mid-wall).
Angle beam examination was carried out in both pulse-echo (reflection) mode as well as pitch-catch (transmission) mode. In the case of pulse-echo examination, the echo signal is generated by the reflected ultrasound from a flaw; the pulse-echo examination was used especially for flaw evaluation, but it also gave good results for detection. The pitch-catch mode was used for detection and for location of flaws (ID, OD, within wall). In pitch-catch mode, the amplitude drop of the ultrasound beam received at the receiving transducer is recorded. It gave very good results, even for flaws with unfavourable orientations.
For the normal beam examinations, flaw detection was based on the amplitude drop and time- of-flight of the first OD backwall echo, or on the amplitude and time-of-flight of the echo reflected by the flaw. Also, in the case of ID surface profiling, the amplitude and time-of- flight of the surface echo are monitored.
1.5.6. Canada
Flaw detection and characterization with UT was performed with four shear wave transducers and one normal beam probe assembled in a cluster. A 20 MHz normal beam transducer was also used for surface profiling of the ID flaws.
Flaw detection and characterization with ET was performed with two surface probes: the first, a Ghent 1 (G1) [11], is designed to compensate for lift-off noise, thereby improving signal-to-
noise for detection of flaws; and the second, a transmit-dual receive (TDR) probe is used to differentiate between axially and circumferentially oriented flaws. Both the G1 and TDR probes satisfy the primary detection and sizing requirements for a PT inspection. ET was used to detect and size ID flaws only.
Both ET and UT use sophisticated data acquisition and analysis software. ET results combined with information obtained from UT provide increased confidence in interpretation of PT inspection data [12].
Profilometry was performed on all ID flaws, as is the case for an in-service inspection.
Profilometry was also used as the examination technique to provide full characterization of all flaws (OD and ID) for three samples.
All these techniques are well established and have been used in-service for many years.
2. FLAW DETECTION AND CHARACTERIZATION
2.1. Flaw description in PT samples
All samples were made from un-irradiated Zr-2.5% Nb pieces of PT. Most samples were 103.4 mm ID and 112 mm OD (standard size of a PT), and approximately 500 mm long (except ROM2). A punch mark was made very close to one end of each sample. The punch mark indicates the 12 o’clock position and also identifies the face of the tube for making all the measurements. All samples contain flaws on ID and OD surfaces. One sample (ROM2) also contained flaws within wall. Once the flaws were machined, a cover was installed and sealed on the outside to hide the OD flaws. A length of approximately 50 mm of PT was left uncovered at both ends of both samples to facilitate holding the sample in the fixtures during the inspection. No flaws were machined in this zone of 50 mm on either end of the samples, except for ROM2 in which few flaws were machined, from the PT end, in the wall of the PT.
The samples were supposed to contain the four calibration notches (axial and circumferential ID and OD notches) as well as the three normal beam calibration slots as per the Canadian Standard. The artificial flaws were machined to be as representative as possible to real flaws encountered in reactor.
Dimensions of all calibration features and flaws were obtained either after manufacturing or at the end of the round-robin exercise to provide the flaw truth for comparison with the various NDE techniques used by the participating countries. Flaw truth was determined by profilometry or measuring microscope.
2.1.1. Argentinean sample (ARG1)
The details of flaws in ARG1 sample with respect to their location, orientation, size and characteristics are given in Table 1.
Table 1. Flaw details in ARG1 sample Flaw Size
Flaw
# Location and
Orientation Length
(mm) Width
(mm) Depth (mm)
Characteristics
1 ID, axial 6.1 0.3 0.136 calibration slot
2 OD, axial 6.15 0.4 0.14 calibration slot
3 ID, circumferential 6.2 0.3 0.152 calibration slot 4 OD, circumferential 6.6 0.4 0.16 calibration slot 5 OD, axial 1.4 0.65 0.7 flat-bottom calibration slot 6 OD, axial 1.45 0.7 2.25 flat-bottom calibration slot 7 OD, axial 1.5 0.75 3.75 flat-bottom calibration slot 8 OD, axial 34.5 0.55 0.145 long and shallow notch 9 OD, circumferential 29.2 0.5 2.05 long and deep curved notch
10 OD, round 2.5 2.5 1.522 dent/flat-bottom
11 OD, round 2.35 2.35 0.53 dent/flat-bottom
12 OD, round 2.35 2.35 2.03 dent/flat-bottom
13 OD, round 2.2 2.2 1.035 dent/flat-bottom
14 OD, circumferential 23.45 0.85 1.45 long and deep curved notch 15 OD, axial 38.25 0.6 2.4 long and deep oblique notch 16 OD, circumferential 27.6 0.45 1.56 long and deep oblique notch 17 OD, circumferential 36.6 0.4 2.95 long and deep notch 18 OD, circumferential 19.55 0.4 0.81 long and deep notch 19 ID, circumferential 10 2.1 1.938 long and deep notch
20 ID, axial 3.9 1 0.938 short and deep notch
The method employed to get the flaw truth was direct profilometry for the OD flaws and measurement of the replicas using a microscope for the ID flaws.
Axial OD calibration slotCircumferential OD calibration slotThree flat-bottom calibration slots Long and shallow axial OD notchLong and deep axial OD notch Long and deep circumferential OD curved notchLong and deep circumferential OD Notch
Dents on the OD FIG. 1. Photographs of some of the OD features machined in the ARG1 sample.
Figure 1 shows photographs of some of the OD flaws present in the ARG1 sample.
2.1.2. Indian sample (IND1)
The details of flaws in IND1 sample with respect to their location, orientation, size and characteristics are given in Table 2.
Table 2. Flaw details in IND1 sample Flaw Size
Flaw
#
Location and
Orientation Length (mm)
Width (mm)
Depth (mm)
Characteristics 1 OD, Oblique at 45˚ 5.84 0.21 0.16 DHC 2 ID, equiaxed 2.50 2.3 1.15 smooth debris fret 3 OD, equiaxed 1.46 0.80 0.63 lap type defect near OD 4 ID, axial 11.50 2.3 1.00 bearing pad fret 5 ID, axial 6.00 0.4 0.22 reference flaw 6 ID, circumferential 6.00 0.3 0.21 reference flaw 7 OD, oblique at 20˚ 5.81 0.24 0.16 DHC
8 OD, equiaxed 1.92 1.27 3.67 lap type defect near ID 9 ID, axial 2.00 1.0 1.10 sharp debris fret 10 OD, circumferential 5.88 0.32 0.11 DHC shallower than
reference flaw 11 OD, axial 5.85 0.28 0.13 reference flaw 12 OD, circumferential 5.76 0.27 0.14 reference flaw 13 OD, circumferential 11.96 0.39 0.23 DHC 14 OD, oblique at 20˚ 5.84 0.17 0.13 DHC
15 OD, equiaxed 1.62 1.15 2.14 lap type defect at mid-wall 16 OD, axial 5.80 0.24 0.068 DHC shallower than
reference flaw 17 OD, axial 11.88 0.35 0.17 DHC
The dimensions of all the OD flaws were measured using a microscope with an accuracy of 0.01 mm for length and 0.005 mm for width and depth. The dimensions of all the ID flaws were taken from the profilometry results reported by Canada.
Figure 2 shows photographs of some of the OD and ID flaws present in the IND1 sample.
OD axial reference notchOD circumferential reference notchOD oblique 20º, same size as reference notch OD oblique 45º, same size as reference notch OD equiaxed, 0.63 mm deepOD equiaxed, 2.14 mm deep OD equiaxed, 3.67 mm deepID sharp debris fret ID bearing pad fret FIG. 2. Photographs of some of the OD and ID features machined in the IND1 sample.
2.1.3. Korean sample (KOR1)
The details of flaws in KOR1 sample with respect to their location, orientation, size and characteristics are given in Table 3.
Table 3. Flaw details in KOR1 sample Flaw Size
Flaw
#
Location and
Orientation Length
(mm) Width
(mm) Depth (mm)
Characteristics
1 OD, axial 6.0 0.3 0.47 short notch deeper than calibration slot
2 OD, tilted 6.0 1 0.93 short and deep notch
3 OD, oblique 6.5 0.3 1.01 short and deep notch
4 OD, circumferential 6.0 0.3 0.41 short notch deeper than calibration slot
5 OD, tilted 6.0 1.0 0.98 short and deep notch
6 OD, oblique 6.5 0.3 1.04 short and deep notch
7 OD, axial 20.0 5.2 0.26 fretting mark
8 ID, circumferential 6 0.3 0.29 short notch deeper than calibration slot
9 OD, oblique 6.0 0.4 1.07 short and deep notch
10 ID, circumferential 6.0 0.3 0.23 short notch similar in depth to calibration slot
11 ID, axial 25.5 2.1 0.22 fretting mark
12 ID, oblique 6.5 0.2 0.23 short notch similar in depth to calibration slot
The flaw truth for the KOR1 sample was provided by Canada through profilometry.
Figure 3 shows photographs of some of the ID flaws present in the KOR1 sample. Pictures of unintentional scratches and debris frets are included for interest.
ID axial bearing pad fretID circumferential reference notch ID oblique 45º Unintentional ID scratchesUnintentional ID debris fret with lift-off Unintentional ID debris fret with lift-off FIG. 3. Photographs of some of the ID features present in the KOR1 sample.
2.1.4. Romanian sample (ROM2)
The details of flaws in ROM2 sample with respect to their location, orientation, size and characteristics are given in Table 4.
Table 4. Types of flaws contained in ROM2 sample PT Ring
Code
Ring Height (mm)
Type of
Artificial Flaws Flaw Orientation
1. Numbe r of Flaws
round-bottom hole axial 16
flat-bottom hole radial 2
through-wall hole radial 1
EDM channel axial 2
ROM-2-H 12
EDM channel circumferential 2
round-bottom hole radial 21
milling channel circumferential 2 ROM-2-R 50
milling channel axial 2
milling channel axial 7
milling channel axial 4
ROM-2-A 14.3
cylindrical long hole axial 1
Figures 4 and 5 show the ROM2 sample with the three rings.
FIG. 4. Outside view of the ROM2 sample.
FIG. 5. Inside view of the ROM2 sample.
Stainless steel tube 112.4 mm OD, 4.3 mm thick
and 120 mm long
Stainless steel sleeve for fastening and centering of PT rings containing artificial flaws
Stainless steel tube 112.4 mm OD, 4.3 mm thick
and 50 mm long. This part contains the punch mark
PT ring ROM-2-A (112.4 mm OD, 4.3 mm thick and 14.3 mm high)
PT ring ROM-2-R (112.4 mm OD, 4.3 mm
thick and 50 mm high) PT ring ROM-2-H (112.4 mm OD, 4.3 mm
thick and 12 mm high)
2.1.4.1. Type and dimensions of flaws in the ROM2 sample
The ROM2 sample contained the following families of flaws (see enclosed CD for all dimensions and further details about the ROM2 sample):
(I) Radial holes of the following types
(I.1) Round-bottom holes machined on the ROM-2-R ring by drilling with conical-end drill, from the outside of the PT. Optical examination in longitudinal cross-section of holes of this type, machined in the same conditions, demonstrated that their bottom is rounded.
Various depths were obtained: 0.4 mm, 0.8 mm and 1.2 mm.
(I.2) Flat-bottom holes achieved by EDM from the inside and outside of the PT. There were 2 flaws of this type on the ROM-2-H PT ring, with the same dimensions 1 mm diameter and 0.5 mm in depth.
(I.3) Through-wall hole machined by auger drilling. The ROM-2-H PT ring contained one such flaw with a 1 mm diameter.
(II) Axial and circumferential channels of the following types
(II.1) Flat-bottom channels machined by EDM. There were 4 flaws of this type on the ROM-2-H PT ring, two in the axial direction, and two in the circumferential direction, one of each type on ID and OD surfaces. All these flaws are 5 mm long, 0.2 mm wide and 0.1 mm deep.
(II.2) Round-bottom (U-form) channels machined by milling. There were 4 flaws of this type on the ROM-2-R PT ring, two in the axial direction on the OD, and two in the circumferential direction (one on the ID and one on the OD). All these channels were about 12 mm long (in the constant depth zone), 0.2 mm wide and 0.1 mm deep. Also, there were seven milling U-form channels, machined with depths between 0.15 mm and 3 mm from the OD on the ROM-2-A PT ring; all these channels are 0.25 mm wide.
(II.3) Angled (V-form) channels machined by milling. There were 4 flaws of this type on the ROM-2-A PT ring.
(III) Axial holes
These were flaws machined in the PT wall by drilling with conical-end auger in the axial direction, from the OD surface. Because of the conical-end auger, the bottom of these holes is rounded. There were 16 flaws of this type on the ROM-2-H PT ring, eight on each ring face.
(IV) Axial long hole
There was only one hole of 1 mm diameter and 14.3 mm length machined by drilling at mid- wall on the ROM-2-A PT ring.
Figure 6 shows the location of most flaws present in the ROM2 sample. This is an UT C-scan image obtained with the normal beam first backwall amplitude while performing the detection scan.
2.1.4.2. Dimensional characterization
As a general remark on flaw dimensions, depending on the machining technology used (auger drilling, milling or EDM), it was observed that the real flaw dimensions were very closed to the dimensions of the tools used (especially for the hole diameters and channel widths).
FIG. 6. Flaw location in the ROM2 sample.
The machined flaws were dimensionally characterized by direct optical examination in the case of small depth OD flaws, and by replica measurements in the case of ID flaws and deep OD flaws.
The real depth of the flaws, (shown in the enclosed CD), was verified, also, by accurate time- of-flight UT measurements, using a USIP12 flaw detector and the DTM12 time-of-flight module. The accuracy of these measurements is ±1 ns (or, equivalent, ±0.005 mm) in the AVERAGE mode, if the temperature is controlled within ±0.5oC.
2.1.5. Chinese sample (CHI1)
The details of flaws in CHI1 sample with respect to their location, orientation, size and characteristics are given in Table 5.
The flaw truth for the CHI1 sample was provided by Canada through profilometry.
Figure 7 shows photographs of some of the OD flaws present in the CHI1 sample.
#4 #1
#2 #7 #8 #9 #10 #11 #12
#14 #15 #16 #17 #18 #19 #20
#31 #32 #33 #34 #35 #36 #37
#21 #23
#24
#25 #26 #28 #29 #30
#38 #39 #40 #41 #44 #45
#53 #54 #55 #56 #57 #58 #59
ROM-2-A ring of 14.3 mm high ROM-2-R ring of 50 mm high
ROM-2-H ring of 12 mm high
Table 5. Flaw details in CHI1 sample Flaw Size
Flaw
# Location and
Orientation Length
(mm) Width
(mm) Depth (mm)
Characteristics
1 OD, axial 5.0 0.5 0.58 short narrow notch
2 OD, axial 6.0 0.3 0.18 short narrow notch
3 OD, axial 12.5 0.6 0.76 long narrow notch of
variable depth 4 OD, circumferential,
tilted 6.0 0.3 0.38 short notch of variable depth
5 OD, circumferential 6.0 0.4 0.6 short notch 6 OD, circumferential 5.0 0.3 0.21 short shallow notch 7 OD, cross like 6.0 0.3 0.29 short cross-like notch
8 OD, axial 5.5 0.3 0.22 short shallow notch
9 ID, axial 6.0 0.3 0.13 short shallow notch
10 ID, axial 6.0 0.2 0.15 short shallow notch
11 ID, axial 6.0 0.3 0.19 short shallow notch
12 OD 10.5 5.3 0.18 tilted square large and
shallow 13 OD, circumferential 8.0 0.3 0.13 long shallow notch 14 OD, circumferential 8.0 0.2 0.16 long shallow notch 15 OD, circumferential 8.0 0.3 0.15 long shallow notch 16 OD, circumferential 8.0 0.2 0.26 long shallow notch
17 OD, axial 5.5 0.2 0.15 short shallow notch
18 OD, axial 5.0 0.5 0.51 short narrow notch
19 OD, axial 5.0 0.4 0.63 short narrow notch
20 OD, circumferential 5.0 0.2 0.17 short shallow notch 21 OD, circumferential 5.5 0.4 0.29 short notch 22 OD, circumferential 5.0 0.3 0.71 short notch
23 ID, axial 5.5 0.2 0.15 short shallow notch
24 ID, circumferential 5.0 0.2 0.14 short shallow notch 25 OD, oblique 6.0 0.2 0.16 short shallow notch 26 OD, oblique 6.0 0.2 0.15 short shallow notch
OD short axial slot OD short circumferential slot OD short oblique slot OD short oblique slot OD long axial slot Two parallel OD circumferential slots OD cross-like indication OD rectangular indication FIG. 7. Photographs of some of the OD features machined in the CHI1 sample.
2.1.6. Canadian samples (CAN1 and CAN2)
The details of flaws in CAN1/2 samples with respect to their location, orientation, size and characteristics are given in Table 6 and Table 7.
Table 6. Flaw details in CAN1 sample Flaw
# Location Length1 (mm)
Width2 (mm)
Depth3
(mm) Characterization E1 OD 1.07 0.22 0.124 axial DHC (semi-elliptical) E2 OD 1.05 0.199 0.28 axial DHC (semi-elliptical) E3 OD 1.98 0.214 0.125 axial DHC (semi-elliptical) E4 OD 2.05 0.204 0.27 axial DHC (semi-elliptical) F1 OD 0.163 1.1 0.125 circumferential DHC (semi-elliptical) F2 OD 0.157 1.06 0.265 circumferential DHC (semi-elliptical) F3 OD 0.148 1.94 0.12 circumferential DHC (semi-elliptical) F4 OD 0.146 2.04 0.27 circumferential DHC (semi-elliptical) C1 ID 3.32 3.24 ~1.42 complex ~axial debris fret of 3 depths
C2 ID 3.9 2.15 0.995
0.834
simple axial debris fret with axial DHC (semi-elliptical)
C3 ID >3.2 3.3 1.49
1.49
complex ~axial debris fret with axial DHC (semi-elliptical)
C4 ID 4.04 2.14 ~0.78 simple axial debris fret
D1 ID 2.3 1.7 0.47 circumferentially-oriented undercut simple debris fret
D2 ID 1.7 2.3 ~0.85
0.73
axially-oriented undercut simple debris fret with axial DHC (semi-elliptical) D3 ID 1.64 2.3 0.71 axially-oriented undercut simple debris fret
D4 ID 2.3 1.83 ~0.85
0.68
circumferentially-oriented undercut simple debris fret with axial DHC (semi-elliptical) J1 OD 19.5 ~0.1 ~0.08 axial installation scratch
Table 7. Flaw details in CAN2 sample Flaw
# Inside/
Outside Length
(mm) Width
(mm) Depth
(mm) Characterization C1 OD 1.93 0.18 0.31 axial DHC (semi-elliptical) B1 ID 12.4 3.25 0.23 simple bearing pad fret
B2 ID 14.4 4.65 0.2495
0.152 double overlapping bearing pad frets
B3 ID 12.4 3.32 0.428
0.251
simple bearing pad fret with angled DHC (semi-elliptical) in corner
B4 ID 14.3 4.72
0.4266 0.249 0.145
double overlapping bearing pad frets with angled DHC (semi-elliptical) in corner
Flaw truth was obtained by profilometry. Figure 8 shows photographs of some of the OD and ID flaws present in the CAN1/2 samples.
5 First depth is deepest fret. Second depth is shallower fret.
l calibration slot OD circumferential calibration slot Normal beam calibration slot (15%) ial DHCOD circumferential DHC Axial debris fret with axial DHC bris fret with axial DHC Single bearing pad fret Double overlapping bearing pad frets FIG. 8. Photographs of some of the OD and ID features machined in the CAN1/2 samples.
2.2. Inspection and diagnostics techniques 2.2.1. Argentina
Argentina used UT examinations for detection, location and characterization of flaws in the various PT samples. The shear wave pulse echo technique was employed using four shear wave transducers, two (in opposite directions) in the axial direction more sensitive to circumferential indications and two (in opposite directions) in the circumferential direction more sensitive to axial indications. The four reference notches from the Canadian Standard (ID axial, ID circumferential, OD axial and OD circumferential) were used for calibration and setting of the detection level.
The equipment used was a NEUS 8.80 Multichannel UT instrument. The UT transducers used were four 10 MHz, 38 mm focus, 0.250” diameter. The reference sensitivity was set according to the Canadian Standard.
2.2.2. India
India primarily employed UT examination for detection, location and characterization of flaws in the PT samples. For some of the samples, ET was also employed for detection of ID flaws.
The inspection head comprises six UT transducers, four angle beam and two normal beam.
The angle beam transducers are used in pairs, one looking axially and the other circumferentially. These are positioned in such a way that the beam from one transducer reaches the other after travelling two skip distances in the PT. For normal beam scanning, one of the transducers was focused on the OD, while the other was focused on the ID.
The PT samples were also examined by ET using a separate inspection head. ET was performed to detect the ID flaws. The curvature of the transducer was modified to match the inside surface of the PT and reduce the effect of lift-off variations. The probe wobbling signal is set horizontal by phase adjustment in the equipment. The amplitude of signal and its phase for ID flaws were recorded.
2.2.2.1. UT techniques for flaw detection
The following UT techniques were employed for detection of flaws in PTs during the CRP:
• Angle beam pulse-echo in axial direction (reflection mode);
• Angle beam pulse-echo in circumferential direction (reflection mode);
• Amplitude drop monitoring during angle beam pitch-catch in axial direction (transmission mode);
• Amplitude drop monitoring during angle beam pitch-catch in circumferential direction (transmission mode);
• Amplitude drop monitoring during normal beam examination using 10 MHz transducer focused on OD of PT.
2.2.2.2. Specifications of UT and ET sensors
All the UT transducers used during the examination are spherically focused. The specifications for these transducers and the ET sensor for different types of examination are given below:
• UT angle beam, axial and circumferential scan: 10 MHz, 10 mm diameter, 30 mm focal length;
• UT normal beam OD scan: 10 MHz, 6 mm diameter, 25 mm focal length;
• UT normal beam ID scan: 20 MHz, 6 mm diameter, 12 mm focal length;
• ET self-comparison differential coil, 300 kHz.
2.2.2.3. Reference sensitivity set-up
Prior to examination of PT samples, the reference sensitivity for angle beam examination is set using reference notches. These notches are machined on a separate PT sample on ID and OD surfaces in axial and circumferential directions. The notches are 6 mm long, 0.2 mm wide and 0.15 mm deep. For normal beam examination, scanning sensitivity is set by adjusting the first backwall signal to approximately 80% of full screen height (FSH). For pitch-catch amplitude monitoring, the transmitted signal was set to approximately 80% FSH.
2.2.2.4. Scanning sensitivity
For angle beam pulse-echo examination, scanning was carried out at 12dB above the reference. For normal beam and angle beam pitch-catch, scanning was carried out at the reference sensitivity.
2.2.2.5. Recording
For angle beam pulse-echo scans, any indication exceeding 6dB over noise at the scanning sensitivity (generally, 20% FSH) is evaluated for recording and reporting. This criterion corresponds to a lower recording level (more stringent) than the recording criteria of 20% of reference level.
For normal beam scans, any drop in the amplitude of backwall signal by more than 5 to 10%
of reference level is recorded.
For angle beam pitch-catch scans, any drop in the amplitude of transmitted signal by more than 5 to 10% of reference level is recorded. This scan provided a high degree of reliability for detection of flaws oriented unfavourably (oblique and tilted flaws).
2.2.2.6. NDE techniques for flaw sizing
Once the flaw is detected and recorded, its length, width and depth are sized. The length and width of the flaws are determined using the 6dB drop technique using angle beam and normal beam signals. Following techniques are employed for assessing the depth of flaws:
• Time-of-flight in axial pitch-catch configuration;
• Time-of-flight in circumferential pitch-catch configuration;
• Time-of-flight using 10 MHz transducer for OD flaws;
•
FIG. 9. B-scan images for simulated flaws.
2.2.3. Republic of Korea
Rep. of Korea employed an ET method for the detection and characterization of flaws in PT samples. The ET send-receiver transducer was designed to detect primarily ID flaws and attempt the detection of OD flaws; but this was known to be very challenging due to the penetration limitation of the magnetic field. This ET probe consists of two identical coils: one coil acts as a sender, and the other coil as a receiver. The transducer coil size is 3.5 mm OD, and it contains a ferrite core. For this inspection, 60 kHz was selected as the primary frequency to detect ID flaws and 20 kHz was chosen, as a second frequency, to attempt the detection of the OD flaws. 100 kHz was used as an additional frequency for confirmation.
For signal analysis, A- and C-scan displays were used.
The calibration was performed using 20%, 50%, and 70% through wall notches. The recording criterion was set two times over the reference sensitivity. Any signals over reference level was recorded and evaluated. The correct flaw location was estimated using the encoder counts. The flaw length and width were sized by the 6dB drop technique. The flaw depth was estimated by measuring the phase angle of the signals.
2.2.4. Romania-NIRDTP
During this CRP, Romania-NIRDTP used an ET method for flaw detection and characterization. A new type of transducer, referred to as the RMF send-receive transducer was developed.
For the case of Zr-2.5% Nb (μ=μ0=4π*10-7 H/m and σ=1.89*106 S/m), with a frequency of 40 kHz, the standard depth of penetration is δ = 1.83 mm. The current density on the PT OD surface is ten times smaller than the density on the ID surface, still sufficient to detect material discontinuities on both PT surfaces and within the wall.
The RMF transducer is an ‘absolute’ send-receive type of transducer. The emission part is made from three rectangular coils 120º apart from each other, star connected and fed with a three-phased electrical current. The reception part is made of 24 identical coils arranged into a circumferential array. See Figure 10 for a photograph of the RMF send-receive transducer.
Debris Fret DHC
FIG. 10. Photograph of ET RMF send-receive transducer.
To use such transducer, special control equipment was designed and fabricated, comprising three programmable function generators type AWG 7223 with two outputs each supplying the alternative three-phased current system feeding the emission side of the transducer. The reception coils are sequentially interrogated, the signal being applied to the ET card, SFT6000N type, thus modified to accept an external reference. The interrogation is made using an analog multiplexer. The electronic boards were inserted into a PC. The signals delivered by the ET boards are stored in the computer memory and processed with a Matlab program. Figure 11 presents a schematic of this system.
Due to the weight of the PT samples, they were mounted into an automatic chuck lathe. The transducers were automatically moved using 2 mm steps. Figure 12 presents a photograph of the experimental set-up.
FIG. 12. Experimental set-up of the ET inspection system used at NIRDTP.
During this CRP, theoretical aspects of the proposed transducer were studied using dyadic Green’s functions and volume integral methods. The results of this work are presented in several publications [1–9], [13–16]. The model developed allows the solving of the forward problem to predict the response of the RMF transducer for different types of discontinuities, placed both on ID and OD surfaces of the PT [6]. At the same time, the inverse problem was solved to provide a method for evaluation of the geometrical shape and dimensions of the flaws from the response of the transducer [8].
The RMF send-receive transducer has shown very good detection capability for flaws situated on both ID and OD surfaces of the PT. The transducer’s spatial resolution (minimum distance between two discontinuities distinctly distinguished) can be improved by the use of smaller reception coils. In the case of the RMF send-receive transducer, one reception coil covers approximately 15º of the PT circumference. It is thought that the use of many reception coils placed as an array improved the detectability. However, this complicates the hardware and software required and may result in a slower inspection speed.
The advantages of the RMF send –receive transducer used during this CRP are:
- detection of flaws both on the ID and OD surfaces - good scanning speed
The drawbacks of this transducer are:
- poor spatial resolution due to large number of reception coils - difficulty in data interpretation
- difficulty of precise assessment of circumferential position and orientation of flaws 2.2.5. Romania-NNDT
Thorough UT inspections were performed by Romania-NNDT on the various PT samples using the ISIS equipment developed at NNDT. The flaw detection and sizing methodology as
2.2.5.1. Inspection equipment
The ISIS-UT and ET system is a complex vertical laboratory system for simulating fuel channel inspections. The functional structure of ISIS-UT&ET integrates all applicable flaw detection and sizing UT and ET methods addressed by the Canadian Standard. This equipment consists of the following three main modules:
(1) Independent Scanning Inspection System (ISIS)
It is a vertical high resolution scanning system that can be used for PT samples up to 750 mm in length. The maximum resolution of the system is 0.02 mm in the vertical direction and 0.01 mm in the circumferential direction.
(2) Multichannel UT and ET unit
It consists of UT and ET instrumentation specific to each NDE function in a multichannel configuration. Depending on the NDE method or inspection requirements under testing, specific modular examination heads are used, with various transducer instrumentation or different interrogation geometry. Configuration of the UT multichannel system for flaw detection and sizing is as follows:
Normal beam, high resolution – broadband examination. This channel is based on the USIP 12/Krautkramer instrument; for UT spectral investigations, a Tektronix digital oscilloscope (100 MHz bandwidth, 1 GS/s sampling rate) is used.
Standard “W” pitch-catch configuration for shear wave examination, based on the multichannel KB 6000/Krautkramer instrument. Alternatively, a corresponding number of one-channel USIP 12/Krautkramer instruments can be used, as was the case in this CRP.
UT dimensional measurements unit, for measurement of the internal diameter and wall thickness of the PT sample.
ET flaw detection and sizing based on the dual frequency MIZ-17/ZETEC instrument.
(3) Command/Control and Data Acquisition/Processing unit (CC&DAP).
The CC&DAP unit performs the real-time command and control of ISIS, the A/D conversion of the detected flaw signals, the synchronization of the entire system, and the data storage. It is also used for data analysis.
PT samples were examined during separate scans with adequate resolution using normal beam (longitudinal waves) and angle beam (shear waves) pulse-echo and pitch-catch configurations.
The transducers used are spherically focused 10 MHz nominal frequency, 6.35 mm effective diameter, and 35 mm focal distance. These transducers are used for both normal beam and shear wave examinations. Also, high-frequency normal beam examinations were performed with the polymer foil transducer (25 MHz nominal frequency, 25 mm focal distance, and a spot size of about 0.3 mm).
in the clockwise direction. The axial movement of the inspection head is performed going up in the sample under investigation.
2.2.5.3. Examination techniques
A minimum of seven UT examinations was performed with the ISIS system to cover the requirements of the Canadian Standard. These are: normal beam – surface echo, normal beam – flaw echo, normal beam – back echo, pitch–catch (transmission) axial, pitch–catch (transmission) circumferential, shear waves (reflection) axial, shear waves (reflection) circumferential.
Regarding flaw detection, very good results were obtained with the axial and circumferential angle beam pitch-catch amplitude and time-of-flight monitoring, combined with high- resolution C-scan imaging. This technique enables determination of the flaw position (ID, OD, or within wall).
Romania-NNDT also used a 25 MHz normal beam broadband transducer focused on the OD with spectral analysis of the echo signals. This examination technique, based on Romania- NNDT research results [17], has proven to be very effective for sizing of flaws on the OD and within wall. UT based on spectral analysis is believed to have the potential to improve the flaw characterization and sizing capability in comparison with the “classical” broadband amplitude evaluation.
2.2.5.4. Flaw sizing methodology
Dedicated software was developed and used as support for data analysis [18].
The 6dB drop technique using angle beam and normal beam was used to determine the position of a flaw and its length and width. Depth is the most important dimension when sizing flaws. Flaw depth is evaluated by the time-of-flight (TOF) images using normal beam.
However, for flaws not necessarily small but having unfavourable orientation for normal beam reflection, flaw depth determination from the normal beam time-of-flight is not possible. In these cases, the shear wave amplitude comparison method is used. It consists of comparing the shear wave amplitudes of the flaw with a calibration feature of similar dimensions (known depth). Unfortunately, the flaw echo amplitude can be affected by many factors and, therefore, the corresponding depth estimate is not very accurate.
Romania-NNDT also applied the Time-Of-Flight Diffraction (TOFD) technique. This technique uses the modulation of the TOF versus flaw sound path generated by the interference of the elementary waves reflected on different surface elements of the flaw. The flaw depth is determined from the maximum variation ΔTOF.
Flaw sizing was essentially based on the evaluation of the various C-scan images obtained during the different examinations. However, for each PT sample, a number of selected indications were re-evaluated by direct (stationary) signal analysis by optimal positioning on the flaw (especially for verification of the depth values of the near ID surface flaws).
